DLNA is based on UPNP A/V.
UPNP A/V ended up being an interoperability nightmare. The UPNP A/V standard is very open ended. Many features are optional. There is no baseline set of media formats that devices have to support. Video format support is a problem for pretty much all media devices. At the time that UPNP A/V was released, this was particularly true in a time when Apple and Microsoft were actively warring on "standard" media formats, and few vendors were willing to adopt open formats such as FLAC, and MKV whose patent and licensing status was up in the air at the time, or to pony up for an endless laundry list of patent portfolio licenses required to play standardized formats.
In addition, the UPNP A/V standards were very loosely specified. Utterly extraordinary readings of the standard were common. Minimalist implementations were the rule rather than the exception. And the pursuit of minimalism lead to some pretty extraordinary readings of the UPNP A/V standard.
DLNA was an attempt to fix the shortcomings of UPNP A/V by heaping thousands of pages of additional requirements on top of the UPNP A/V standards. The DLNA standards organization provided standardized tests suites that certified devices had to pass.
According to the DLNA specs, DLNA devices SHOULD be compatible with UPNP A/V devices, according to the terms of the DLNA standard. But there is no requirement that they MUST be compatible. So (surprise surprise) more often than not, they are not compatible. In fairness, some common UPNP implementations were so jaw-droppingly bad that this is not entirely the fault of the DLNA consortium.
DLNA also had its problems. It originally cost somewhere in the neighborhood of $50,000 to get a set of the specs, and (tens of?) thousands of dollars to get the certification, and frankly, plus an additional requirement in practice to acquire a bottomless pit of ISO standards documents in the ISO MPEG family of standards at enormous expense, since these standards were incorporated by reference in the DLNA standards, which then in turn incorporate by reference other ISO standards. All at huge expense. All of which in turn precluded any sensible open-source DLNA implementation.
UPNP A/V on the other hand, had been published in freely available documents.
Plus the sheer size of the DLNA specifications, which still had all kinds of CANs and MUSTs and SHOULDs liberally scattered through the requirements. Making it painfully easy for two certified DLNA devices to want to have nothing to do with each other because of incompatible CANs and SHOULDs.
So when it was all said and done, alhtough interoperability had improved dramatically, it still wasn't that great.
Most of the time these days, UPNP devices mostly interoperate with DLNA devices but aren't certified (because of the prohibitive cost of doing so) so they can't be called DLNA devices.